Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method comprising: receiving, by a computing device, a request for media items to recommend to a user of a user device; obtaining, by the computing device, recommendable media items; ranking, by the computing device, the recommendable media items according to affinity scores associated with the user; generating, by the computing device, an adventurousness score for the user that indicates a magnitude of willingness of the user to explore or experience different media content than has been previously consumed by the user, wherein the adventurousness score is based on a churn count associated with the user, wherein the churn count represents a number of times the user has selected a media item associated with a new media item category not previously selected by the user within a time period; calculating, by the computing device, a satisfaction gain score for each media item in the ranked media items based at least in part upon the adventurousness score and the ranking; and selecting and providing, by the computing device, a portion of the recommendable media items to recommend to the user based on the satisfaction gain score calculated for each media item in the ranked media items.
A method for personalized media recommendations improves content suggestions by balancing user preferences with exploration. The system receives a request for media recommendations and retrieves a set of candidate media items. These items are ranked based on affinity scores, which reflect the user's historical preferences. To enhance recommendation diversity, the system calculates an adventurousness score for the user, quantifying their willingness to explore new content. This score is derived from a churn count, which tracks how often the user engages with media from previously unexplored categories within a defined timeframe. The system then computes a satisfaction gain score for each ranked media item, incorporating both the affinity score and the adventurousness score. This hybrid approach ensures recommendations align with user preferences while encouraging exploration. Finally, the system selects and provides a subset of media items to the user based on the highest satisfaction gain scores, optimizing both relevance and novelty. This method addresses the challenge of recommendation systems becoming overly repetitive by dynamically adjusting suggestions to user behavior.
2. The method of claim 1 , wherein the affinity scores are calculated based on the user's previous media item selections and represent the user's preference for media items of a particular category.
This invention relates to personalized media recommendation systems, specifically improving the accuracy of recommendations by calculating affinity scores based on a user's historical media selections. The system identifies patterns in the user's past choices to determine their preference for specific media categories, such as genres, artists, or content types. These affinity scores quantify the user's likelihood of engaging with similar media items, enabling the system to prioritize recommendations that align with their demonstrated preferences. The method enhances recommendation accuracy by dynamically adjusting scores based on ongoing user interactions, ensuring recommendations remain relevant over time. This approach addresses the challenge of delivering personalized content in media streaming platforms, where user preferences may evolve, by continuously refining affinity scores to reflect current interests. The system may also incorporate additional factors, such as contextual data or external trends, to further refine recommendations. The invention aims to improve user satisfaction by providing more tailored and engaging media suggestions.
3. The method of claim 1 , wherein generating the adventurousness score for the user includes: determining one or more characteristics of the user; and grouping the user into a group with other users based on the one or more characteristics of the user.
This invention relates to systems for assessing user behavior, specifically determining a user's adventurousness score based on their characteristics and grouping them with similar users. The technology addresses the challenge of personalizing recommendations or experiences by quantifying how adventurous a user is, which can be applied in fields like entertainment, travel, or e-commerce. The method involves analyzing one or more characteristics of a user, such as their past behavior, preferences, or demographic information, to generate an adventurousness score. This score reflects the user's likelihood to engage in novel or risky activities compared to others. The user is then grouped with other users who share similar characteristics, allowing for targeted recommendations or content tailored to their adventurousness level. The grouping may be based on clustering algorithms, statistical analysis, or machine learning techniques to identify patterns among users with comparable traits. By categorizing users into distinct groups based on adventurousness, the system enables more accurate personalization, improving user engagement and satisfaction. This approach can be used in recommendation engines, marketing strategies, or adaptive interfaces where understanding user tendencies is critical. The invention enhances existing systems by incorporating behavioral and demographic data to refine user segmentation and deliver more relevant experiences.
4. The method of claim 3 , wherein the one or more characteristics include an average amount of time that the user plays media items, an age of the user, a geographic location of the user, or a combination thereof.
This invention relates to personalized media recommendation systems that analyze user characteristics to improve content suggestions. The system collects and processes data about a user's media consumption habits, including the average time spent playing media items, the user's age, and their geographic location. By evaluating these characteristics, the system generates tailored recommendations that align with the user's preferences and behavior. The method involves tracking media playback duration to determine engagement levels, assessing demographic factors like age to refine recommendations, and incorporating geographic data to account for regional preferences. The combined analysis of these factors enhances the accuracy and relevance of media suggestions, ensuring a more personalized user experience. The system dynamically adjusts recommendations based on real-time or historical data, optimizing content delivery for individual users. This approach addresses the challenge of generic media recommendations by leveraging specific user attributes to provide more targeted and engaging content. The invention improves user satisfaction by delivering media that better matches their interests and habits.
5. The method of claim 3 , further comprising: ordering users within the group based on the churn count; and generating the adventurousness score for the user based on a position of the user within the ordered group.
This invention relates to a system for assessing user behavior in a group-based environment, particularly for determining a user's adventurousness score based on their interaction patterns. The problem addressed is the need to quantify how adventurous or exploratory a user is within a group, which can be useful for recommendations, engagement strategies, or personalized content delivery. The method involves tracking user interactions within a group, specifically counting how often a user churns or deviates from typical behavior. This churn count is used to rank users within the group, with higher churn counts indicating more adventurous behavior. The adventurousness score is then calculated based on the user's position in this ranked list, where higher rankings correspond to higher adventurousness scores. This approach allows for a dynamic and comparative assessment of user behavior, enabling systems to adapt to individual preferences and engagement levels. The method may also include additional steps such as identifying a group of users, tracking their interactions, and calculating churn counts for each user. The adventurousness score is derived from the user's relative position in the ordered group, providing a scalable and objective measure of exploratory behavior. This can be applied in various domains, including social networks, e-commerce, or content platforms, to enhance user experience and engagement.
6. The method of claim 1 , wherein calculating the satisfaction gain score for each media item in the ranked media items comprises: determining a number of times a media item associated with a particular media item category has been viewed by the user; providing the number as input to a submodular function for generating a satisfaction gain score for the user, wherein the submodular function is determined based on the adventurousness score; and calculating the satisfaction gain score based on a result of the submodular function.
This invention relates to personalized media recommendation systems that optimize user satisfaction by dynamically adjusting recommendations based on user behavior and preferences. The problem addressed is the challenge of balancing novelty and relevance in media recommendations to prevent user fatigue while maintaining engagement. The method involves calculating a satisfaction gain score for each media item in a ranked list of recommendations. For each item, the system determines how many times the user has viewed media from the same category as the item. This count is then input into a submodular function, which generates a satisfaction gain score tailored to the user's adventurousness. The adventurousness score reflects the user's willingness to explore new content versus sticking to familiar preferences. The submodular function ensures that the recommendation system dynamically adjusts to the user's behavior, promoting diversity when the user is open to new content and reinforcing familiarity when the user prefers consistency. The final satisfaction gain score is derived from the submodular function's output, influencing the ranking of media items to maximize long-term user engagement. This approach improves recommendation accuracy by personalizing the balance between novelty and relevance based on individual user tendencies.
7. The method of claim 1 , comprising determining the satisfaction gain score based on a satisfaction gain for a genre of content.
This invention relates to a system for evaluating user satisfaction with content recommendations, particularly in media streaming or digital content delivery platforms. The problem addressed is the need to accurately measure and improve user satisfaction by analyzing how different genres of content contribute to user engagement and enjoyment. The method involves calculating a satisfaction gain score, which quantifies the positive impact of recommended content on user satisfaction. This score is determined by assessing the satisfaction gain specifically for a genre of content, meaning the system evaluates how much a particular genre enhances user experience compared to other genres. The method may also include tracking user interactions with recommended content, such as viewing duration, ratings, or explicit feedback, to refine the satisfaction gain score over time. Additionally, the system may adjust recommendations based on the satisfaction gain scores of different genres, ensuring that future suggestions align with user preferences and maximize satisfaction. By focusing on genre-specific satisfaction gains, the system improves the accuracy of content recommendations, leading to higher user retention and engagement. This approach helps platforms tailor their offerings more effectively, ensuring that users consistently receive content that aligns with their interests and enhances their overall experience.
8. A non-transitory computer-readable medium including one or more sequences of instructions that, when executed by one or more processors, cause the processors to perform operations comprising: receiving, by a computing device, a request for media items to recommend to a user of a user device; obtaining, by the computing device, recommendable media items; ranking, by the computing device, the recommendable media items according to affinity scores associated with the user; generating, by the computing device, an adventurousness score for the user that indicates a magnitude of willingness of the user to explore or experience different media content than has been previously consumed by the user, wherein the adventurousness score is based on a churn count associated with the user, wherein the churn count represents a number of times the user has selected a media item associated with a new media item category not previously selected by the user within a time period; calculating, by the computing device, a satisfaction gain score for each media item in the ranked media items based at least in part upon the adventurousness score and the ranking; and selecting and providing, by the computing device, a portion of the recommendable media items to recommend to the user based on the satisfaction gain score calculated for each media item in the ranked media items.
This invention relates to personalized media recommendation systems that enhance user engagement by balancing familiarity and novelty in recommendations. The system addresses the challenge of recommending media items that are both relevant to a user's preferences and sufficiently diverse to maintain interest over time. The system receives a request for media recommendations and retrieves a set of candidate media items. These items are initially ranked based on affinity scores, which reflect the user's historical preferences. To encourage exploration, the system calculates an adventurousness score for the user, derived from a churn count—a metric tracking how often the user selects media from new categories within a given timeframe. This score quantifies the user's willingness to engage with unfamiliar content. The system then computes a satisfaction gain score for each ranked media item, incorporating both the affinity score and the adventurousness score. This score predicts how much satisfaction the user will derive from the recommendation, balancing familiarity and novelty. Finally, the system selects and provides a subset of media items to the user based on these satisfaction gain scores, optimizing for both relevance and exploration. The approach ensures recommendations remain engaging by dynamically adapting to the user's evolving preferences and exploratory behavior.
9. The non-transitory computer-readable medium of claim 8 , wherein the affinity scores are calculated based on the user's previous media item selections and represent the user's preference for media items of a particular category.
This invention relates to a system for calculating and utilizing affinity scores to personalize media recommendations. The problem addressed is the challenge of accurately predicting user preferences for media items, such as songs, videos, or articles, based on their past interactions. The system calculates affinity scores that quantify a user's preference for media items within specific categories, such as genres, artists, or topics. These scores are derived from the user's historical selections, such as plays, skips, or ratings, to reflect their likelihood of engaging with similar content. The system then uses these affinity scores to generate personalized recommendations, ensuring that suggested media items align with the user's demonstrated preferences. By dynamically updating the scores as new user interactions occur, the system adapts to evolving tastes over time. This approach improves recommendation accuracy and user satisfaction by prioritizing content that matches the user's established preferences while allowing for gradual exploration of new categories. The invention is implemented via a non-transitory computer-readable medium containing instructions for executing the affinity scoring and recommendation processes.
10. The non-transitory computer-readable medium of claim 8 , wherein the instructions that cause generating the adventurousness score for the user include instructions that cause: determining one or more characteristics of the user; and grouping the user into a group with other users based on the one or more characteristics of the user.
This invention relates to a system for assessing user adventurousness in a digital environment, particularly for personalized content recommendations or user experience customization. The problem addressed is the need to accurately measure a user's willingness to engage with novel or unconventional content, activities, or interactions, which is valuable for platforms like social media, e-commerce, or entertainment services. The system generates an adventurousness score for a user by analyzing their characteristics, such as behavior, preferences, or demographic data. These characteristics are used to group the user with similar individuals, allowing for comparative analysis or collective insights. The grouping may be based on shared traits like risk tolerance, exploration habits, or engagement patterns. By categorizing users into adventurousness-based groups, the system enables tailored recommendations or targeted interventions, improving user satisfaction and platform engagement. The adventurousness score is derived from a combination of explicit and implicit data, such as user interactions with experimental features, feedback on unconventional content, or historical behavior. The grouping mechanism ensures that users are compared against relevant peers, enhancing the accuracy of the adventurousness assessment. This approach supports dynamic adjustments to user experiences, such as recommending bolder content to highly adventurous users or simplifying interfaces for more cautious users. The system operates on a non-transitory computer-readable medium, ensuring scalability and integration with existing digital platforms.
11. The non-transitory computer-readable medium of claim 10 , wherein the one or more characteristics include an average amount of time that the user plays media items, an age of the user, a geographic location of the user, or a combination thereof.
This invention relates to a system for personalizing media recommendations based on user characteristics. The system analyzes user data to generate tailored media suggestions, improving user engagement and satisfaction. The core challenge addressed is the need for more accurate and relevant media recommendations by incorporating specific user attributes. The system processes user data to determine one or more characteristics, such as the average duration a user spends playing media items, the user's age, their geographic location, or a combination of these factors. These characteristics are used to refine media recommendations, ensuring they align with the user's preferences and behavior. The system may also track user interactions with recommended media to further optimize future suggestions. By leveraging these characteristics, the system enhances the relevance of media recommendations, leading to higher user engagement and a more personalized experience. The approach ensures that recommendations are dynamically adjusted based on real-time user data, improving the overall efficiency of media delivery platforms. This method is particularly useful in digital media streaming services, where personalized content is critical for user retention and satisfaction.
12. The non-transitory computer-readable medium of claim 10 , wherein the instructions cause: ordering users within the group based on the churn count; and generating the adventurousness score for the user based on a position of the user within the ordered group.
13. The non-transitory computer-readable medium of claim 8 , wherein the instructions that cause calculating the satisfaction gain score for each media item in the ranked media items include instructions that cause: determining a number of times a media item associated with a particular media item category has been viewed by the user; providing the number as input to a submodular function for generating a satisfaction gain score for the user, wherein the submodular function is determined based on the adventurousness score; and calculating the satisfaction gain score based on a result of the submodular function.
This invention relates to personalized media recommendation systems that optimize user satisfaction by dynamically adjusting recommendations based on user behavior and preferences. The problem addressed is the challenge of maintaining user engagement by balancing familiar content with novel recommendations, particularly for users with varying levels of adventurousness in their media consumption habits. The system calculates a satisfaction gain score for each media item in a ranked list of recommendations. This score is derived by determining how frequently a user has viewed media items within a specific category. The frequency count is then input into a submodular function, which generates a satisfaction gain score tailored to the user's adventurousness score. The submodular function is designed to adapt recommendations based on whether the user prefers familiar content or is open to exploring new categories. The adventurousness score influences the function's weighting, ensuring recommendations align with the user's tendency to seek novelty or stick to known preferences. This approach enhances recommendation relevance and user satisfaction by dynamically adjusting the balance between familiar and novel content.
14. A system comprising: one or more processors; and a non-transitory computer-readable medium including one or more sequences of instructions that, when executed by the one or more processors, cause the processors to perform operations comprising: receiving, by a computing device, a request for media items to recommend to a user of a user device; obtaining, by the computing device, recommendable media items; ranking, by the computing device, the recommendable media items according to affinity scores associated with the user; generating, by the computing device, an adventurousness score for the user that indicates a magnitude of willingness of the user to explore or experience different media content than has been previously consumed by the user, wherein the adventurousness score is based on a churn count associated with the user wherein the churn count represents a number of times the user has selected a media item associated with a new media item category not previously selected by the user within a time period; calculating, by the computing device, a satisfaction gain score for each media item in the ranked media items based at least in part upon the adventurousness score and the ranking; and selecting and providing, by the computing device, a portion of the recommendable media items to recommend to the user based on the satisfaction gain score calculated for each media item in the ranked media items.
The system improves media recommendation by balancing user preferences with exploration of new content. It addresses the challenge of recommending media that aligns with user tastes while encouraging discovery of unfamiliar content. The system uses a computing device with processors and non-transitory storage to process recommendations. Upon receiving a request for media recommendations, the system retrieves recommendable media items and ranks them based on affinity scores reflecting the user's historical preferences. It then calculates an adventurousness score for the user, which quantifies their willingness to explore new content. This score is derived from a churn count, representing how often the user selects media from previously unexplored categories within a defined time period. The system further computes a satisfaction gain score for each ranked media item, incorporating both the adventurousness score and the item's ranking. Finally, the system selects and provides a subset of media items to the user based on their satisfaction gain scores, optimizing for both personal relevance and novelty. This approach enhances recommendation diversity while maintaining user satisfaction.
15. The system of claim 14 , wherein the affinity scores are calculated based on the user's previous media item selections and represent the user's preference for media items of a particular category.
A system for personalized media recommendations calculates affinity scores to determine a user's preference for media items in specific categories. The system analyzes the user's past media selections to generate these scores, which quantify the user's likelihood of engaging with content from particular categories. By leveraging these affinity scores, the system tailors media recommendations to align with the user's demonstrated preferences, improving the relevance and accuracy of suggested content. The system may also incorporate additional factors, such as user behavior patterns or contextual data, to refine the recommendation process. This approach enhances user satisfaction by delivering more personalized and engaging media suggestions, addressing the challenge of providing relevant content in an increasingly diverse and expansive media landscape. The system dynamically updates affinity scores as the user interacts with more media, ensuring recommendations remain aligned with evolving preferences. This method improves the efficiency of content discovery and consumption, benefiting both users and media providers by optimizing engagement and retention.
16. The system of claim 14 , wherein the instructions that cause generating the adventurousness score for the user include instructions that cause: determining one or more characteristics of the user; and grouping the user into a group with other users based on the one or more characteristics of the user.
This invention relates to a system for assessing user adventurousness in a social or interactive platform. The system evaluates a user's willingness to engage in novel or risky activities by generating an adventurousness score. This score is derived from analyzing user behavior, preferences, and interactions within the platform. The system also categorizes users into groups based on shared characteristics, such as demographics, past activities, or engagement patterns, to refine the adventurousness assessment. By grouping users, the system can tailor recommendations, content, or experiences to better match individual or group preferences, enhancing user engagement and satisfaction. The adventurousness score may influence personalized suggestions, such as travel destinations, social connections, or interactive challenges, ensuring a more dynamic and relevant user experience. The system leverages data-driven insights to dynamically adjust recommendations, fostering a more personalized and engaging platform environment.
17. The system of claim 16 , wherein the one or more characteristics include an average amount of time that the user plays media items, an age of the user, and a geographic location of the user.
This invention relates to a system for personalized media recommendations, addressing the challenge of delivering relevant media content to users based on their preferences and behaviors. The system collects and analyzes user data to generate tailored recommendations, improving user engagement and satisfaction. Specifically, the system identifies one or more characteristics of the user, such as the average amount of time the user spends playing media items, the user's age, and the user's geographic location. These characteristics are used to refine the recommendation process, ensuring that suggested media aligns with the user's habits, demographic factors, and regional preferences. The system may also incorporate additional user-specific data, such as listening or viewing history, to further enhance the accuracy of recommendations. By leveraging these insights, the system dynamically adjusts the selection and presentation of media content, providing a more personalized and engaging experience. The invention aims to optimize media consumption by adapting to individual user behaviors and preferences, ultimately increasing user retention and satisfaction.
18. The system of claim 16 , wherein the instructions cause: ordering users within the group based on the churn count; and generating the adventurousness score for the user based on a position of the user within the ordered group.
This invention relates to a system for assessing user behavior in a group-based environment, particularly focusing on predicting user churn and determining user adventurousness. The system analyzes user interactions within a group to identify patterns indicative of churn, which refers to the likelihood of a user leaving or disengaging from the group. By tracking and quantifying these patterns, the system assigns a churn count to each user, representing their propensity to churn. The system then orders users within the group based on their churn counts, allowing for a comparative analysis of user behavior. Additionally, the system generates an adventurousness score for each user based on their position within the ordered group. This score reflects the user's willingness to engage in novel or risky activities, with higher scores indicating greater adventurousness. The adventurousness score is derived from the user's relative position among peers, providing insights into their behavior relative to the group's overall dynamics. This approach enables targeted interventions to retain users at risk of churn and tailor experiences to match individual adventurousness levels.
19. The system of claim 14 , wherein the instructions that cause calculating the satisfaction gain score for each media item in the ranked media items include instructions that cause: determining a number of times a media item associated with a particular media item category has been viewed by the user; providing the number as input to a submodular function for generating a satisfaction gain score for the user, wherein the submodular function is determined based on the adventurousness score; and calculating the satisfaction gain score based on a result of the submodular function.
This invention relates to a media recommendation system that personalizes content suggestions based on user behavior and preferences. The system addresses the challenge of recommending media items that balance user familiarity with novel content, ensuring engagement while avoiding repetitive suggestions. The system calculates a satisfaction gain score for each media item in a ranked list of recommendations. This score is derived from the frequency with which a user has viewed media items within a specific category. The system inputs this frequency into a submodular function, which generates the satisfaction gain score. The submodular function is tailored to the user's adventurousness score, a metric that reflects their willingness to explore new content. By incorporating this score, the system dynamically adjusts recommendations to either favor familiar content or introduce novel items, depending on the user's preference for exploration. The system also includes a method for ranking media items based on multiple factors, such as relevance, novelty, and user engagement history. The ranked list is then refined using the satisfaction gain score to ensure recommendations align with the user's adventurousness level. This approach enhances user satisfaction by providing a personalized balance between familiar and new content.
20. The system of claim 14 , wherein the churn count represents a number of times the user has viewed, for at least a threshold amount of time, the media item associated with the new media item category not previously viewed by the user within the time period.
21. The system of claim 20 , wherein the threshold amount of time comprises at least twenty seconds.
A system for monitoring and managing user interactions with a computing device includes a sensor module configured to detect user presence and activity, such as touch, motion, or biometric signals. The system also includes a processing module that analyzes the detected activity to determine whether the user is actively engaged with the device. If the user remains inactive for a threshold period of at least twenty seconds, the system triggers a security or power-saving response, such as locking the device, dimming the display, or initiating a sleep mode. The system may also include a notification module to alert the user before transitioning to a low-power state. The processing module may adjust the threshold based on contextual factors, such as application usage or environmental conditions, to balance security and usability. The system is designed to prevent unauthorized access and conserve energy by automatically managing device states based on detected inactivity.
Unknown
July 14, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.